Method of discriminating pictures

• US 4,741,046 A
• Filed: 07/22/1985
• Issued: 04/26/1988
• Est. Priority Date: 07/27/1984
• Status: Expired due to Fees

• Current Assignee: Konishiroku Photo Industry Co., Ltd.
• Original Assignee: Konishiroku Photo Industry Co., Ltd.
• Inventors: Emori, Yasufumi, Yasuda, Yoshizumi, Matsunawa, Masahiko
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Couso, Jose L.
• Application Number:
  US06/757,653
• Time in Patent Office: 1,009 Days
• Field of Search: 382/9, 382/18, 382/48, 382/51, 382/41, 382/49, 382/54, 358/282, 358/283, 358/280, 358/296, 358/298
• US Class Current: 382/176
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

A method of discriminating pictures, wherein a picture is binary-coded. The variation in the number of foreground pixels constituting foreground pixel regions within a discrimination unit area is determined. Alternatively the variation in the distance between said foreground pixel regions is determined. It is then determined by a known relationship whether the discrimination unit area is of a dot gradation picture or not. If the picture is not a dot gradation picture, the rate of background is measured to determine whether the above discrimination unit area is of a line picture or of a continuous tone picture.

Claims

1. A method of discriminating whether a picture is a halftone dot picture, a continuous tone picture or a line picture, said picture containing at least one discrimination unit area therein, each said at least one discrimination unit including a plurality of foreground pixel regions, each of which regions consists of any one foreground pixel and any additional foreground pixels which are immediately adjacent to the one foreground pixel or any continuously adjacent pixels in one of the major directions, said method comprising the steps of:

a. binary-coding the picture using a predetermined number of pixels, thereby determining whether each pixel of the picture is a foreground pixel or a background pixel;

b. determining the statistical variation in at least one of the following;

(1) the number of foreground pixels constituting foreground pixel regions within a discrimination unit area; and

(2) the distance between said foreground pixel regions within a discrimination unit area;

c. determining whether or not said discrimination unit area is part of a halftone dot picture from a known relation between said variation and whether said picture is a halftone dot picture; and

d. if it is determined that the picture is not a half-tone dot picture, then determining from a ratio of background pixels to foreground pixels whether or not said discrimination unit area is of a line picture, according to a known relationship.

2. A method according to claim 1, wherein said discrimination unit areas are obtained by dividing the picture into a plurality of smaller areas.

3. A method according to claim 1, wherein

the picture is divided into a plurality of blocks, there being the possibility of connections between said blocks,

each block is sorted into prescribed patterns by using the binary coding of the picture,

any connections between blocks are checked on the basis of said patterns so as to extract picture regions, and

the extracted picture regions are used as the discrimination unit areas.

4. A method according to claim 3, wherein any data obtained by projecting two-dimensionally the binary coding of the picture of each of said plurality of blocks are used for sorting said blocks into patterns.

5. A method according to claim 3 or 4, wherein when the connections between said blocks are checked, each of the blocks is subdivided into four equal small blocks, and any connections between said small blocks and all small blocks surrounding them are checked.

6. A method according to claim 1, wherein

the picture is divided into a plurality of blocks,

the picture is sorted into prescribed patterns by using the binary coding of the picture, and

the frequency of occurrence of the patterns is plotted to determine the type of picture.

7. A method of discriminating whether a picture is a halftone dot picture, a continuous tone picture or a line picture, said picture containing at least one discrimination unit area therein, each said at least one discrimination unit including a plurality of foreground pixel regions, each of which regions consists of any one foreground pixel and any additional foreground pixels which are immediately adjacent to the one foreground pixel or any continuously adjacent pixels in one of the major directions, said method comprising the steps of:

a. determining the statistical variation in at least one of the following;

(1) the number of foreground pixels constituting foreground pixel regions within a discrimination unit area; and

(2) the distance between said foreground pixel regions within a discrimination unit area;

b. determining whether or not said discrimination unit area is part of a halftone dot picture from a known relation between said variation and whether said picture is a halftone dot picture.